红外与激光工程, 2005, 34 (2): 216, 网络出版: 2006-05-25
基于知识表示与有监督学习的动态红外图像分析
Dynamic infrared imagery analysis method based on knowledge representation and supervised learning
动态图像分析 运动物体检测 有监督学习 知识表达 场景建模 Dynamic image analysis Moving object detection Supervised learning Knowledge representation Scene modeling
摘要
动态图像分析是一个令人感兴趣的研究课题,广泛地应用于交通监测、场景监控和预警等方面.由于红外图像的噪声相对较大,对比度不高,因此红外序列图像的动态分析有着自己的特点.提出了一种基于知识表示与有监督学习的动态红外图像分析方法,能有效地利用序列图像全局建模进行红外序列图像中的运动物体检测、特性判别行为分析.
Abstract
Dynamic image analysis is an interesting research field, and it is widely used in traffic monitoring, scene surveillance and warning forecast. Dynamic analysis of infrared imagery sequence has its character because of its high noise levels and low contrast.A dynamic infrared imagery analysis method based on knowledge representation and supervised learning is proposed. This algorithm can efficiently apply global modeling of sequential images to moving objects detection, pattern discrimination and behavior analysis in infrared imagery sequences.
张蔚, 张天序, 沈俊. 基于知识表示与有监督学习的动态红外图像分析[J]. 红外与激光工程, 2005, 34(2): 216. 张蔚, 张天序, 沈俊. Dynamic infrared imagery analysis method based on knowledge representation and supervised learning[J]. Infrared and Laser Engineering, 2005, 34(2): 216.